Aster Adds 'data-Application Server' in Version 4.0

The new "data-application server" provides major performance gains, since information no longer needs to be transmitted over a network to a separate application server, according to Aster.

The approach is enabled by an framework that integrates SQL with the MapReduce programming model first developed at Google for parallel processing of large data sets across pools of commodity hardware.

The IT industry has entered a world where multi-terabyte and even petabyte-scale analytics workloads are a reality, but traditional data pipelines aren't ready to handle the job, said Aster CEO Mayank Bawa. "In order to do large-scale analytics on large data sets, you have to bring the application to the data itself," he said.

Packaged applications or custom ones developed in Java, C, C++, C#, .NET, Perl and Python run inside resource containers in Aster's system, with no need to rewrite any code, Bawa said. Aster will support additional languages in the future, including R, the open-source analytics language, he said.

Aster's database can handle hundreds of concurrent workloads, which are managed and prioritized through a new rules-based system. Data and applications are co-located in the database but have separate supporting services, improving overall stability.

It is not especially difficult to work with the new system, and Aster isn't expecting to rack up major dollars for new services engagements, according to Bawa. "Services is a good side business, but people should be able to implement it very easily with some side training," he said.

Aster is also in discussions with a number of global systems integrators that want to build practices around the database, according to a spokeswoman.

Once customers get more acquainted with the new architectural approach, it will enable an entirely new set of applications, Bawa predicted.

Aster's announcement is "a slick vision" with "a not-so-slick" name, analyst Curt Monash said in a blog post.

"Of course, Aster is hardly the only analytic DBMS vendor to have the idea of explicitly enhancing general analytic processing; that's why we see lots of MapReduce announcements," he added. "But I don't know of anybody else whose approach is quite so elegant and general at this time."

Other observers echoed Monash. "It's really innovative and I don't use those terms lightly," said Forrester Research analyst James Kobielus. Moving application logic into the data warehousing environment is "a logical next step" that has implications not just for BI and analytics, but transactional business applications as well, he said.